DEVELOPMENT OF OBSTACLE CROSSING SYSTEM FOR UNMANNED AERIAL VEHICLE

Abstract

Most existing unmanned aerial vehicles (UAV) are piloted using remote controls. However, when performing some works (geological, geodetic, rescue, exploration) there are restrictions related to the insufficient working range of the UAV. Therefore, there is a need for automatic control of the UAV. When automatically controlling the UAV, one of the tasks is to form the trajectory of the UAV, when performing geodetic, search and other works, including automatic maneuvering to bypass obstacles. In the absence of information on the exact location of obstacles, it is advisable to manage on the basis of fuzzy logic. The article presents a scheme of fuzzy inference. The set of values of fuzzy input linguistic variables (LV), output LV forms a fuzzy database, the set of rules of fuzzy products forms a fuzzy knowledge base. Each fuzzy number is represented by triangular or trapezoidal ordered real numbers and is given by the membership function. Input signals of sensors are used to describe the space in the direction of UAV movement. After bypassing the obstacle, the UAV can be navigated using GPS. The space in the direction of the UAV is divided into five sectors, which analyze the presence of obstacles and the distance to them. Three input linguistic variables were introduced - two for the analysis of the location of the obstacle (LV "Horizontal" and LV "Vertical") and one for the analysis of the distance to the obstacle (LV " Distance"). Regarding the output variables, depending on the location of the obstacle, the UAV will change the yaw angle, speed and height. Three source drugs were introduced - "Angle", "Speed" and "Height" and their membership functions were built. The base of fuzzy production rules of a kind is made: if "values of input variables ", then "values of output variables". Mamdani fuzzy inference system is used. The Rule Viewer and the Surface Viewer Fuzzy Logic Toolbox analyzed the results of the fuzzy system. The approach developed in this article to prevent UAV collisions with obstacles based on the fuzzy logic apparatus differs from the known analogues in the way of UAV coordination, a set of control rules. The use of fuzzy logic allows you to successfully solve the problem of forming the trajectory of the UAV depending on the location of the obstacle and the distance to it.